Analysis of Genomic Aberrations and Gene Expression Profiling Identifies

Analysis of genomic aberrations and gene expression proﬁling identiﬁes novel lesions and pathways in myeloproliferative neoplasms

KL Rice1, X Lin1, K Wolniak1, BL Ebert2, W Berkofsky-Fessler3, M Buzzai4, Y Sun5,CXi5, P Elkin5, R Levine6, T Golub7, DG Gilliland8, JD Crispino1, JD Licht1 and W Zhang5

1Division of Hematology/Oncology, Department of Medicine, Northwestern University Feinberg School of Medicine, Chicago, IL, USA; 2Division of Hematology, Department of Medicine, Brigham and Women’s Hospital, Boston, MA, USA; 3Translational Research Sciences, Hoffmann-La Roche, Inc., Nutley, NJ, USA; 4Oncology, Novartis, Origgio, VA, Italy; 5Division of Hematology/ Oncology, Department of Medicine, Center of Biomedical Informatics, Mount Sinai School of Medicine, New York, NY, USA; 6Human Oncology and Pathogenesis Program and Leukemia Service, Department of Medicine, Memorial Sloan Kettering Cancer Center, New York, NY, USA; 7The Eli and Edythe L. Broad Institute of Massachusetts Institute of Technology and Harvard University, Cambridge, MA, USA and 8Oncology, Merck Research Laboratories, North Wales, PA, USA

Polycythemia vera (PV), essential thrombocythemia and by a recurrent mutation, JAK2V617F, which is present in B95% primary myelofibrosis, are myeloproliferative neoplasms of patients with PV, B65% with PMF and B55% with ET.2 JAK2 (MPNs) with distinct clinical features and are associated with is one member of the Janus family of non-receptor tyrosine the JAK2V617F mutation. To identify genomic anomalies involved in the pathogenesis of these disorders, we profiled kinases that transduces signals downstream of type I and II 87 MPN patients using Affymetrix 250K single-nucleotide cytokine receptors via signal transducer and activators of polymorphism (SNP) arrays. Aberrations affecting chr9 were transcription (STATs), and is an important regulator of cell the most frequently observed and included 9pLOH (n ¼ 16), survival, proliferation, differentiation and apoptosis.3,4 The trisomy 9 (n ¼ 6) and amplifications of 9p13.3–23.3 (n ¼ 1), JAK2V617F mutation leads to autophosphorylation of JAK2 9q33.1–34.13 (n ¼ 1) and 9q34.13 (n ¼ 6). Patients with trisomy and constitutive activation of downstream pathways including 9 were associated with elevated JAK2V617F mutant allele burden, suggesting that gain of chr9 represents an alternative the Janus kinase (JAK)/STAT, phosphatidylinositol 3 kinase/AKT mechanism for increasing JAK2V617F dosage. Gene expres- (v-akt murine thymoma viral oncogene homolog 1) and sion profiling of patients with and without chr9 abnormalities mitogen-activated protein kinase (MAPK)/extracellular signal- ( þ 9, 9pLOH), identified genes potentially involved in disease regulated kinase pathways.5–9 Although overexpression of pathogenesis including JAK2, STAT5B and MAPK14. We also JAK2V617F in a murine bone marrow transplant models leads observed recurrent gains of 1p36.31–36.33 (n ¼ 6), 17q21.2– to a PV-like disease and megakaryocytic hyperplasia, such mice q21.31 (n ¼ 5) and 17q25.1–25.3 (n ¼ 5) and deletions affecting 18p11.31–11.32 (n ¼ 8). Combined SNP and gene expression do not develop thrombocytosis and questions regarding the analysis identified aberrations affecting components of a non- effect of allele dosage on disease phenotype remain unan- canonical PRC2 complex (EZH1, SUZ12 and JARID2) and genes swered. The finding of Theocharides et al.10 that 70% of comprising a ‘HSC signature’ (MLLT3, SMARCA2 and PBX1). JAK2V617F-positive MPN patients went on to develop We show that NFIB, which is amplified in 7/87 MPN patients and JAK2V617F-negative acute myeloid leukemia (AML), suggests upregulated in PV CD34 þ cells, protects cells from apoptosis that additional predisposing factors may precede the JAK2 induced by cytokine withdrawal. mutation. Recently, several acquired inactivating mutations in Blood Cancer Journal (2011) 1, e40; doi:10.1038/bcj.2011.39; 11 (ref. 12) (ref. 13) published online 11 November 2011 TET2, ASXL1 and IDH1/2 in MPN progenitors Keywords: myeloproliferative neoplasms; JAK2V617F; NFIB; SNP were identified, and it is likely that additional genes are involved in the initiation of a pre-leukemic clone or as disease modifiers, in co-operation with JAK2V617F. Previous studies using standard karyotyping revealed abnorm- Introduction alities in 50% of PMF patients, 15% of PV and o5% of ET patients.14–16 Although such conventional chromosome analysis The Ph-negative myeloproliferative neoplasms (MPNs) including has proven useful for diagnosis and prognostic information in polycythemia vera (PV), essential thrombocythemia (ET) and PMF patients and other myeloid malignancies, the sensitivity of primary myelofibrosis (PMF) are clonal hematopoietic disorders these approaches is limited. Indeed, studies by Tefferi using characterized by the overproduction of blood cells belonging to comparative genomic hybridization (CGH) oligonucleotide the myeloid, megakaryocytic and/or erythroid lineages.1 arrays to identify genomic aberrations in a panel of MPN Although clinically distinct, these diseases are characterized patients, detected copy number aberrations (CNAs) in 29% and 18% of patients with PV and ET, respectively, while cytogenetics for these patients appeared normal.17 Recently, high- Correspondence: Dr W Zhang, Division of Hematology/Oncology, Department of Medicine, Center of Biomedical Informatics of Mount resolution single-nucleotide polymorphism (SNP) arrays have Sinai School of Medicine, Mount Sinai School of Medicine, One emerged as a powerful, high-throughput technology to allow Gustave L Levy Place, New York, NY 10029, USA. detection of micro-deletions and duplications and inference of E-mail: [email protected] loss of heterozygosity (LOH) from genotype calls, in cases where or Professor JD Licht, Division of Hematology/Oncology, Department conventional cytogenetic technologies and array CGH are of Medicine, Robert H Lurie Comprehensive Cancer Center, unsuccessful. Since then, SNP arrays have been successfully Northwestern University Feinberg School of Medicine, 303 East Superior Street, Lurie 5-123, Chicago, IL 60611, USA. used to identify cryptic oncogenic deletions and duplications E-mail: [email protected] affecting putative oncogenes and tumor suppressors in various Received 16 June 2011; accepted 11 August 2011 hematological malignancies. For example, the identification of Identification of novel lesions and pathways in MPN KL Rice et al 2 abnormalities on chromosome 4q24 affecting the TET2 gene in then used to determine the JAK2V617F allele burden (%) in patients with MDS, MPN and AML was facilitated by use of SNP DNA extracted from individual patient granulocytes. array analysis.11,18 Similarly, PAX5 was identified as a target for mutation in B-cell acute lymphocytic leukemia using SNP arrays.19 Recent studies using high-resolution SNP arrays Data analysis identified novel, recurrent aberrations in a large cohort of CNA analysis. These data were processed and analyzed MPN patients and revealed associations between specific using a custom developed procedure (Supplementary Figure S1). chromosomal anomalies and disease progression.20 Genotype calls were generated by GTYPE (Affymetrix) from CEL Here, we report an integrative analysis of chromosomal files and 250K library files for Nsp arrays. Raw data were abnormalities and gene expression changes in MPN patients. normalized using invariant set normalization method. The We profiled CNAs in peripheral blood granulocytes of 87 intensity value for each SNP was extracted using a model-based patients with MPN and identify new regions affected by perfect match/mismatch algorithm in dChip (http://biosun1. recurrent chromosomal aberrations. In a subset of the profiled harvard.edu/complab/dchip/). The log2 ratio of the intensity patients, genomic copy number changes could be correlated for each SNP for each sample relative to data from 60 with gene expression in granulocytes, leading to the identifica- unrelated CEU subjects of HapMap project (http://www. tion of novel genes deregulated in MPN. We show that patients hapmap.org) was calculated. The circular binary segment affected by chromosome 9 abnormalities, which represent the algorithm implemented in BioConductor DNAcopy package most frequent CNA, are characterized by deregulation of set of (http://bioconductor.org) was applied to the log2 ratio data to genes including JAK2, STAT5B and MAPK14, suggesting that segment the chromosome into contiguous regions of equal copy these genes may be involved in disease pathogenesis. We also number based on a maximal t-statistics with the use of a performed meta-analysis of five independent MPN expression permutation reference distribution (using coordinates of the SNP data sets to identify genes commonly deregulated in MPN. based on Genome Build HG-17).22 Segmental value was Comparison of genes identified in these meta-analysis with calculated by averaging copy number change values for the those affected by CNAs, revealed genes that may have important SNPs residing in that segment. The baseline value of genome- roles in disease pathogenesis including EZH1, MLLT3, SOCS3 wide segments in the sample with a bias in regional changes (in and SMARCA2. Finally, we show that the transcription factor case of the samples with more regional gains than losses, or vice NFIB, which is located adjacent to JAK2 and upregulated in our versa or with only regional gains or losses) may be shifted MPN meta-analysis, leads to increased net cell growth and because of the preceding steps of global normalization and decreased apoptosis in the murine hematopoietic cell line, median centering. To correct the baseline, the segmental values Ba/F3-EPOR. were adjusted by subtracting the median of segmental change values for all the SNPs. Considering a large segment with low- level copy number change as important as a small segment with Patients and methods high-level change, the adjusted segmental change value was further weighted by multiplying log10 of the length of that Patient samples segment. A segment in the MPN sample was considered A total of 87 MPN patients (32 PV, 24 PMF and 31 ET) were significant if its segmental value is 43 s.d. of segmental values included in this study (Supplementary Table S1). Samples were in a range of Q13*(Q2Q1) to Q2 þ 3*(Q2Q1) where Q1 or obtained with informed consent at the Dana–Farber Cancer 8 Q2 are the 25% or 75% percentile of segmental values in all the Institute as described previously. samples, respectively. The significant segments having an overlap of 475% with CNA regions in normal subjects reported SNP arrays in the Database of Genomic Variations (http://projects.tcag.ca/ DNA was extracted from peripheral blood granulocytes using variation/) were filtered out for further analysis. The frequencies the QIAamp DNA Blood Maxi Kit (Qiagen, Valencia, CA, USA) of copy number changes in either direction for significant and analyzed with GeneChip Mapping 250K arrays according to segments among MPN samples were summarized for each the manufacturer’s protocol (Affymetrix, Santa Clara, CA, disease using the boundary information of significant segments. USA).11 Briefly, 250 ng of genomic DNA was digested by Nsp To assess the significance of genomic aberrations in a group of restriction enzymes and ligated with the appropriate adaptor for samples, the permutation was performed to assess the frequency amplification with Platinum Pfx DNA polymerase (Invitrogen, of genomic aberrations occurring in the genome by a random Carlsbad, CA, USA). Three 100 ml PCR reactions were carried chance. The genome and aberrant regions were fragmented into out for each adaptor-ligated DNA sample. PCR products were 50 kb blocks and the fragmented blocks for each sample were pooled and purified. In all, 40 mg of PCR product was randomly assigned to any block in the genome. The frequency fragmented with DNase I and visualized on a 4% TBE agarose of aberrations occurring in every block in the genome at random gel. Fragmented PCR products were subsequently end-labeled was calculated and compared with the actual frequency of with biotin. Hybridization and washing was performed using the aberrations. These steps were repeated 1000 times. If the Affymetrix Fluidics Station 450 and Gene chip and chips were random chance of obtaining a proportion higher than actual scanned on an Affymetrix GeneChip Scanner G7 (Affymetrix). frequency of genomic aberrations for a given region is o5% (equivalently to 50 times from 1000 permutations), the recurrent aberration for a region was determined to be significant. The JAK2V617F allele burden analysis Fisher’s exact test was performed to examine whether a The JAK2V617F allele status was assessed using a quantitative particular chromosomal aberration was associated with JAK2 reverse transcriptase-PCR assay as previously described.21 In mutation status or disease type. To determine if the genomic brief, a standard curve for the proportion of JAK2V617F/total instability profiles were segregated by disease type, hierarchical JAK2 against the dCT (Ct-JAK2V617F-CtJAK2WT) was generated clustering of genome-wide copy number change profiles was using DNA from a healthy control and a PV erythroid colony performed in PV, ET and PMF patients using the distance of homozygous for the JAK2V617F mutation. This method was Pearson correlation co-efficiencies.

Blood Cancer Journal Identification of novel lesions and pathways in MPN KL Rice et al 3 LOH analysis. LOH analysis was performed comparing Viral constructs and transduction MPN genotypes with 60 CEU unrelated normal controls derived Full-length human NFIB complementary DNA (cDNA) was from the HapMap project (http://www.hapmap.org). A Hidden obtained from Open Biosystems (Huntsville, AL, USA) and Markov-based algorithm was used to infer LOH from unpaired cloned into the two promoter EF1a-cytomegalovirus-green samples implemented dChip.23 The regions with a minimum of fluorescent protein (GFP) lentiviral vector (Gift of Dr Linzhao 80% probability of LOH call were identified for each sample Cheng, John Hopkins). Full-length human JAK2WT and and LOH regions for MPN and HapMap normal samples were JAK2V617F cDNA was subcloned from the MSCV-Neo/Puro then summarized and compared. retroviral vector8 into the two promoter A2-GFP lentiviral vector. Viral particles were produced by transient transfection of 293T cells with pRSV-REV (Rous sarcoma virus-REV), pMDLg/ RNA extraction and microarray pRRE (gag/pol elements) and pMD.G (envelope elements) Total mRNA was extracted from peripheral blood granulocytes plasmids with Fugene6:DNA (Roche, Mannheim, Germany). of 38 selected MPN patients and 11 normal control patients After 24 h, transfection reagent was removed and replaced with 1 using Trizol (Invitrogen). In all, 10 mg was hybridized to RPMI and incubated at 32 C for 48 h. Supernatants were pooled and filtered through a 0.45mm syringe filter for immediate use. HGU133A V2 arrays (Affymetrix). Data were analyzed using 6 Genespring 11.0.2 (Agilent Technologies Inc., Santa Clara, CA, For transduction of Ba/F3-EPOR or TF-1 cells, 2 10 cells were USA) following RMA normalization and genes with a fold pelleted and resuspended in 2 ml of lentivirus with polybrene m 1 change of X1.5 (Po0.05) were considered statistically sig- (4 g/ml) and spinoculated at 1010 g for 2 h at 25 C. Infected nificant. For MPN meta-analysis, a one-sided t-test (Po0.05) cells were then cultured for 48 h and the percentage of GFP was was performed between disease and control samples for data determined by flow cytometry. sets 1, 2, 3 and 5 (PV data set 1: CD34 þ cells from 10 PV and 9 normal controls,24 PMF data set 2: CD34 þ cells from 3 PMF and 3 normal controls,25 ET data set 3: platelets from 6 ET and 5 cDNA preparation and quantitative reverse normal controls,26 ET data set 4: CD34 þ derived megakar- transcriptase- PCR yocyte cells from 2 ET and 2 normal controls,27 MPN data set 5: cDNA was synthesized from 1 mg of total RNA using Iscript granulocytes from 93 MPN patients and 11 normal controls, cDNA Synthesis Kit (Bio-Rad, Hercules, CA, USA). Quantitative unpublished data, includes 38 MPN patients described above). reverse transcriptase-PCR reactions were performed using For data set 4, an average of log ratios of two samples (X2-fold) Quantitect SYBR Green PCR Kit (Qiagen) using the Mx3000P was used due to limited sample size. Genes deregulated in the Real-Time PCR System (Stratagene, Agilent Technologies, same direction (average of X1.5-fold and present in at least 3/5 La Jolla, CA, USA). Quantitect primers (Qiagen) used for data sets) were considered significant. For analysis of gene quantitative quantitative reverse transcriptase-PCR are located expression between patients with chromosome 9 aberrations in Supplementary Table S2. and normal chromosome 9 cytogenetics, a one-sided t-test was performed and genes with a fold change X1.5 (Po0.05) were considered significant. Cytokine withdrawal Sorted GFP þ Ba/F3-EPOR cells were cultured in RPMI plus erythropoeitin (EPO) (1 U/ml) and cumulative cell growth counts were performed over 10 days by Trypan blue exclusion. For EPO Bioinformatic analysis withdrawal studies, cells were washed three times in phosphate- Gene biofunctions and network analysis was performed using buffered saline and seeded at 1 105 cells/ml with varying Ingenuity Pathway Analysis (Ingenuity Systems, Redwood City, concentrations of EPO (1, 0.1 and 0.01 U/ml) for 2 days. The CA, USA). percentage of dead cells was calculated by Trypan blue exclusion.

Cell culture 293T cells were grown in Dulbecco’s modified Eagle’s medium Results containing 10% (v/v) fetal bovine serum (FBS), 50 U/ml penicillin and 50 mg/ml streptomycin. Ba/F3 cells expressing Identification of novel MPN-associated disease genes the murine erythropoietin receptor (Ba/F3-EPOR) were pre- using a combination of SNP and gene expression viously described28 and cultured in RPMI containing 10% profiling (v/v) FBS supplemented with recombinant erythropoietin hEPO To identify genes recurrently affected by genomic aberrations in (1 U/ml) (Cell Sciences, Canton, MA, USA). HEL cells were MPN, we hybridized DNA from peripheral blood granulocytes grown in RPMI containing 10% (v/v) FBS. TF-1 cells were of 87 MPN patients (32 PV, 24 PMF and 31 ET) to Affymetrix cultured in RPMI containing 10% (v/v) FBS supplemented with 250K SNP arrays. Analysis of genomic profiles for CNAs and human granulocyte-macrophage colony-stimulating factor (2 ng/ml) copy neutral LOH/uniparental disomy confirmed previously (Peprotech, Rocky Hill, NJ, USA). UKE-1 cells (Gift of Gabi reported chromosomal abnormalities in MPN, including recur- Vohwinkel, University Hospital Eppendorf, Hamburg, Germany) rent gain of 1q (n ¼ 4), trisomy 8 (n ¼ 2), 9pLOH (n ¼ 16), were grown in Iscove’s modification of Dulbecco’s medium trisomy 9 (n ¼ 6) and deletions in 20q11.21–13.13 (n ¼ 4)20,29 containing 10% (v/v) FBS, 10% (v/v) horse serum and hydro- (Figure 1a; Supplementary Table S1). We also found that several 6 cortisone (10 M) (Sigma-Aldrich, St Louis, MO, USA). For JAK2 genes mutated in MPN were also affected by genomic inhibition studies, HEL and UKE-1 cells were treated with aberrations in our cohort. For example, analysis of the TET2 either JAK inhibitor I (Calbiochem, Gibbstown, NJ, USA) or gene, which is mutated in B12% of patients with MPN,11 NVP-BSK805 Inhibitor (1–2 mM) (Novartis, Schaumberg, IL, revealed small deletions at chr4q24 affecting TET2 in two PV USA), respectively, for 16 h, using dimethylsulphoxide as a patients (Supplementary Figure S1). We also observed loss of control. 20q11.21–q13.12 in one PV (s328) and one PMF patient (s212),

Blood Cancer Journal lo acrJournal Cancer Blood 4 peuae nMNptet SplmnayTbeS) otm eoi rfieo hoooe6i n ain s4)soigagi t6p22.3 at gain a showing (s242) patient one in 6 chromosome of profile genomic Bottom: S6). Table encompassing (Supplementary patients MPN in upregulated ain s3)soigagi t1q112.1ecmasn 4 ee including genes 144 ( encompassing patient single 17q21.1–21.31 a at represent lines gain Individual a MPN. with showing patients (s531) 87 patient in arrays 250K Mapping 1 Figure Nsi P ( MPN in CNAs JAR1D2 e ie ersn h enlog mean the represent lines Red . a dnicto fnvllsosadptwy nMPN in pathways and lesions novel of Identification .Iega ersnigrgoso hoooa an rd n os(re)ietfiduigAfmti GeneChip Affymetrix using identified (green) loss and (red) gains chromosomal of regions representing Ideogram ).

Log2Ratio Log2Ratio -1.0 -0.5 -1.0 -0.5 0.0 0.5 0.0 0.5 2 ai fteitniyo h ape eaiet 0urltdHpa omlcontrols. normal HapMap unrelated 60 to relative samples the of intensity the of ratio LRice KL tal et EZH1 b .Tp eoi rfieo hoooe1 none in 17 chromosome of profile genomic Top: ). and Chr17 Chr6 ITGA2B hc r lotranscriptionally also are which , Identification of novel lesions and pathways in MPN KL Rice et al 5 Table 1 SNP arrays identify regions of recurrent gain or loss in MPN patients

Chr Ab Start End Dist Freq COSMIC database MPN vs control MPN meta-analysis (kb) (41.5-fold, Po0.05) (41.5-fold, Po0.05)

1 Gain 1p36.33 1p36.31 4338 6/87 PRDM16 LRRC47 F 1 Gain 1q32.1 1q44 45 208 4/87 ELK4, MDM4, ADSS, ANGEL2, ARF1, ARID4B, AKT3, GALNT2, LGALS8, SLC45A3 BTG2, CNIH4, FAIM3, FBXO28, MTR, PSEN2, RAB3GAP2, HEATR1, LGALS8, NUAK2, PFKFB2, RPS6KC1, SMYD2 RAB3GAP2, RCOR3, TOMM20, TRAF3IP3, TSNAX, ZC3H11A 8 Gain 8p23.1 8p23.1 428 4/87 FF F 9 Gain 9p24.3 9p23 10 489 6/87* JAK2 JAK2, SMARCA2 JAK2, SMARCA2 9 Gain 9p23 9p13.3 16 366 7/87* CDKN2A, MLLT3, F ADFP, MLLT3, NFIB NFIB 9 Gain 9p13.3 9q33.1 83 625 6/87* FANCC, FANCG, GNAQ, AGTPBP1, ANXA1, CTNNAL1, ABCA1, CDC14B, DAPK1 NR4A3, OMD, PAX5, DCTN3, DDX58 KLF4, NDUFB6, DDX58, KIAA0367, NTRK2, PTCH, SYK, TAL2 XPA NOL8 NR4A3, SMU1, SPTLC1 T TMOD1, TPM2, TXNDC4 DRD7, TLN1 9 Gain 9q33.1 9q34.13 10 771 7/87* CEP1 CDK5RAP2, FBXW2, GAPVD1, PTGS1, RABGAP1 GOLGA1, PBX3 POLE3, PRPF4, PTGS1, RABGAP1, STOM 9 Gain 9q34.13 9q34.13 1097 10/87* FFANGPTL2 9 Gain 9q34.13 11 137 12/87* ABL1, BRD3, FNBP1, ASB6, CDK9, GPR107, LCN2, F NOTCH1, NUP214, RALGDS SET, SPTAN1, SSNA1 SETTSC1 12 Gain 12p13.33 12p13.33 1008 4/87 FF F 15 Loss 15q15.3 15q15.3 172 7/87 FF F 17 Gain 17q21.2 17q21.31 3984 5/87 BRCA1, ETV4 ATP6V0A1, EZH1, FMNL1 HEXIM1, EZH1, ITGA2B ITGA2B, MAP3K14, PSME3 17 Gain 17q25.1 17q25.3 9770 5/87 ASPSCR1, CANT1 GGA3, MAFG, SEC14L1, SGSH, BAIAP2, EXOC7, SLC16A3 ST6GALNAC2,WBP2 SEC14L1 SOCS3 18 Loss 18p11.32 18p11.31 813 8/87 FF F 20 Loss 20q11.23 20q13.12 9917 4/87 MAFB, TOP1 BLCAP, BPI, CTNNBL1, C20orf67, EYA2, HNF4A MMP9, SFRS6 L3MBTL, SRC, TGM2, UBE2C 20 Gain 20q13.33 20q13.33 3353 4/87 SS18L1 LIME1, PCMTD2, RPS21 DATF1 Abbreviations: CNA, copy number aberration; COSMIC, Catalogue of Somatic Mutations in Cancer; ET, essential thrombocythemia; MPN, myeloproliferative neoplasm; PMF, primary myelofibrosis; PV, polycythemia vera; SNP, single-nucleotide polymorphism. Frequently occurring CNAs identified using 250K SNP Array Analysis in MPN patients (n ¼ 87). Asterisks indicate that these frequencies include six cases of whole chromosome 9 gain. Genes affected by CNA, somatic mutations in cancer (COSMIC) and gene expression analysis (MPN vs normal; genes deregulated in peripheral blood granulocytes from 37/87 MPN patients (10 PV, 17 ET and 10 PMF) compared with 11 normal controls (X1.5-fold, Po0.05; Supplementary Table S4) and meta-analysis of five MPN array data sets (one-sided t-test, Po0.05, average of 1.5-fold in 3/5 data sets; Supplementary Table S6) are shown in the far right columns. a region including the ASXL1 gene, which was recently shown In order to further investigate the disease potential of these to be mutated in B11% of myelodysplastic syndromes (MDS) genomic gains and losses, we compared genes affected by CNAs and in B7% of patients with MPN ET and PMF,30 gain of in our MPN cohort (X2 patients; 4448 genes) (Supplementary ch11q23.3 encompassing CBL in one ET patient (s403) and loss Table S3) with genes deregulated in peripheral blood granulocytes of chr7q35 harboring EZH2 in one PMF patient (s212) from a subset of these patients (n ¼ 37, 10 PV, 17 ET and (Supplementary Table S1). 10 PMF, Supplementary Table S1) compared with normal Overall, our analysis identified CNAs in 88% of MPN controls (n ¼ 11) (X1.5-fold, Po0.05; 1447 genes, Supplemen- patients, (77/87), however, the majority (B90%) of recurrent tary Table S4) and cancer-associated genes obtained from the aberrations occurred at a frequency of o5%. The most frequent Catalogue of Somatic Mutations in Cancer (COSMIC) (426 abnormalities involved chromosome 9, including 9pLOH genes) (Supplementary Figure S2). Of the 426 COSMIC genes, (n ¼ 16) as well as gains of whole chromosome 9 (n ¼ 6) and 75 were affected by CNAs and 42 were deregulated in partial amplifications including 9p13.3–23.3 (n ¼ 1), 9q33.1– expression (Supplementary Table S5). Nine genes fulfilled all 34.13 (n ¼ 1) and 9q34.13 (n ¼ 6). Recurrent gains of 1p36.31– criteria and these included JAK2 and Translocated Promoter 36.33 (n ¼ 6), 17q21.2–q21.31 (n ¼ 5), 17q25.1–25.3 (n ¼ 5) Region, which were both affected by copy number gains and and deletions of 18p11.31–11.32 (n ¼ 8) and 15q15.3 (n ¼ 7) overexpressed in MPN patients compared with normal controls. were also observed (Table 1). In addition to these larger Using this tri-level analysis, we identified aberrations affecting aberrations, a subset of CNAs identified in our screen involved components of a non-canonical PRC2 complex including smaller regions of recurrent gain or loss, typically containing upregulation of SUZ12 (1.8-fold, Po0.05), amplification o10 genes. We hypothesized that these areas may contain and upregulation of EZH1 (5/87; 1.6-fold, Po0.05) and putative oncogenes or tumor suppressors, as in the case of the amplification of JARID2 (2/87) (Figure 1b). Several genes TET2 tumor suppressor. Such candidates include FLI1 and ETS1, constituting part of a HSC ‘gene signature’31 were also identified located in a 754–841-kb region of loss (3/87), TCBA1 located in in our analysis, including MLLT3 and SMARCA2, amplified as a a 600-bp region of loss (3/87) and CNTNAP2 located in a 35-kb result of trisomy 9 in six patients and PBX1 amplified in region of loss (2/87) (Supplementary Table S1). three patients on chr1q22–1q25.3, strengthening the hypothesis

Blood Cancer Journal Identiﬁcation of novel lesions and pathways in MPN KL Rice et al 6 Table 2 MPN-associated pathways Table 3 MPN-associated bio-functions

Pathway P-value Genes Aberration Fold change Functions P-value Genes Aberration Fold change JAK/STAT signaling 3.4 104 AKT3 Gain (3/87) 1.8 JAK2 Gain (6/87) 1.9 Hematological disease (4.7 105–1.2 102) SOCS3 Gain (6/87) 1.9 Erythrocytosis 7.6 104 JAK2 Gain (6/87) 1.9 JAK3 Gain (2/87) 1.5 SOCS3 Gain (6/87) 1.9 IL-8 signaling 4.4 104 AKT3 Gain (3/87) 1.8 SRC Loss (4/87) 1.6

ANGPT1 Gain (2/87) 2.8 3 AZU1 Gain (3/87) 1.5 Leukemia 1.7 10 AKT3 Gain (3/87) 1.8 IKBKB Gain (2/87) 1.6 JAK2 Gain (6/87) 1.9 PTK2B Gain (2/87) 1.8 JAK3 Gain (2/87) 1.5 SRC Loss (4/87) 1.6 MTR Gain (3/87) 1.9 Erythropoietin 5.5 104 AKT3 Gain (3/87) 1.8 PBX1 Gain (3/87) 1.7 signaling JAK2 Gain (6/87) 1.9 PDE4C Gain (2/87) 2.7 SOCS3 Gain (6/87) 1.9 PTGS1 Gain (7/87) 2.1 SRC Loss (4/87) 1.6 SRC Loss (4/87) 1.6 Abbreviations: IL, interleukin; JAK, Janus kinase; MPN, myeloproli- Cell growth and proliferation (1.5 105–1.2 102) ferative neoplasm; SNP, single-nucleotide polymorphism; SOCS, 4 suppressor of cytokine signaling; STAT, signal transducer and Colony 9.2 10 IFI16 Gain (3/87) 1.7 activators of transcription. formation UBE2C Loss (3/87) 1.9 Ingenuity pathway analysis was performed using genes commonly CLU Gain (2/87) 2.0 identified by SNP array (n ¼ 87) and MPN meta-analysis (n ¼ 114) ELF3 Gain (3/87) 1.6 (89 genes). JAK2 Gain (6/87) 1.9 LAPTM4B Gain (2/87) 2.0 PDLIM2 Gain (2/87) 2.8 SRC Gain (4/87) 1.6 that aberrant self-renewal may underlie a common MPN progenitor (Table 1). Proliferation 1.3 103 AKT3 Gain (3/87) 1.8 Finally, we compared genes affected by CNAs with those of cells ANGPT1 Gain (2/87) 2.8 deregulated in a comprehensive MPN meta-analysis. Analysis of BLZF1 Gain (3/87) 1.6 CD244 Gain (3/87) 2.2 five individual MPN data sets comparing patient and normal HRPT2 Gain (3/87) 2.2 control samples identified 574 upregulated genes and 636 CLU Gain (2/87) 2.0 downregulated genes. Comparison of this data set with genes CTSB Gain (2.87) 2.7 affected by genomic aberrations in our MPN patient cohort ELF3 Gain (3/87) 1.6 (CNAs present in X2 patients), identified 81 upregulated genes in ENPP2 Gain (2/87) 1.5 regions of chromosome gain and 20 downregulated genes in HNF4A Loss (3/87) 1.8 HPGD Gain (2/87) 2 regions of loss (Supplementary Table S6). Ingenuity pathway IFI16 Gain (3/87) 1.7 analysis of these genes revealed an enrichment of genes IKBKB Gain (2/87) 1.6 belonging to various signaling pathways including the JAK/STAT, ITGA2B Gain (5/87) 2.3 interleukin-8 and erythropoietin pathways, reinforcing the role of JAK2 Gain (6/87) 1.9 dysregulated JAK/STAT signaling in MPN (Table 2). Furthermore, JAK3 Gain (2/87) 1.5 genes previously implicated in hematological disease such as MAPRE Loss (2/87) 1.5 NFIB Gain (7/87) 2.3 PBX1, PDE4C, PTGS1 and SRC as well as in cell growth and PBX1 Gain (3/87) 1.7 proliferation and apoptosis were also enriched (Table 3). Taken SMARCA2 Gain (6/87) 2 together, these data indicate that genomic aberrations present in SOCS3 Gain (6/87) 1.9 MPN account for a proportion of gene expression changes, which SRC Loss (4/87) 1.6 may have a role in disease pathogenesis. TGM2 Loss (4/87) 3.3 UBE2C Loss (3/87) 1.9

Aberrations involving chromosome 9 are associated Cell death (2.5 106–1.3 102) with dysregulated gene expression Apoptosis 3.2 104 ANGPT1 Gain (2/87) 2.8 Previous studies show that wild-type JAK2 can exert a dominant- CLU Gain (2/87) 2.0 DAPK1 Gain (6/87) 1.8 negative effect over mutant JAK2V617F, suggesting that there DNAJB9 Loss (2/87) –1.6 may be a selective advantage for elevating the ratio of IKBKB Gain (2/87) 1.6 32 JAK2V617F/JAK2WT. We observed that the majority of patients PTK2B Gain (2/87) 1.8 with a high JAK2V617F allele burden (485% JAK2V617F) exhibit SOCS3 Gain (5/87) 1.9 chr9pLOH (86%; 12/14), confirming previous studies that Abbreviations: IL, interleukin; JAK, Janus kinase; MPN, myeloproli- JAK2V617F mutant allele status can be increased through LOH ferative neoplasm; SNP, single-nucleotide polymorphism; SOCS, because of mitotic recombination8 (Supplementary Table S1). We suppressor of cytokine signaling. also found that the majority of patients with trisomy 9 had an Ingenuity pathway analysis was performed using genes commonly elevated JAK2V617F allele burden (average 67%), consistent with identified by SNP array analysis (n ¼ 87) and meta-analysis (n ¼ 114) duplication of the mutant allele. (89 genes). As trisomy 9 and 9pLOH represent the predominant CNAs in MPN, we used gene expression profiling to investigate the downregulated genes that were differentially expressed between differences between patients with normal (n ¼ 25) versus these two groups (X1.5-fold, Po0.05) (Supplementary Table S7). abnormal chromosome 9 cytogenetics (n ¼ 13). We identified Among those genes upregulated in patients with chromosome 9 a set of 493 genes, including 210 upregulated and 283 aberrations, were JAK2, STAT5B, MAPK14 and CD177 (Figure 2).

Blood Cancer Journal Identiﬁcation of novel lesions and pathways in MPN KL Rice et al 7 Chr9+ 9pLOH Chr9 normal

-30 3

Figure 2 Chromosome 9 aberrations are associated with elevated JAK2V617F allele burden and gene expression signature. Gene expression differences in patients with normal (n ¼ 25) versus abnormal ( þ 9, 9pLOH) chromosome 9 cytogenetics (n ¼ 13) (Supplementary Table S7). A subset of genes upregulated (42.7-fold, Po0.05; 26 genes) or downregulated (43-fold, Po0.05; 26 genes) in patients with chromosome 9 anomalies compared with normal chromosome 9 cytogenetics are displayed in a heatmap.

PMF and ET patients display higher frequencies of more moderate sized (1–10 Mb) chromosomal aberrations than chromosomal aberrations compared with PV PV patients (P ¼ 0.01). Although PV, ET and PMF patients may all harbor the JAK2V617F allele, their clinical manifestations differ. This may be at least in part explained by the gene dosage hypothesis, which is based on the observation that homozygous JAK2V617F NFIB overexpression is associated with chromosome 9 mutant erythroid colonies are found in the majority of patients amplification and protects hematopoietic cells from with PV but rarely in ET patients.33 This suggests that higher cytokine withdrawal-induced apoptosis kinase activity (JAK2V617F homozygosity) gives rise to an Accumulating evidence suggests that additional genetic lesions erythroid/PV phenotype, whereas lower kinase activity cooperate with JAK2V617F to induce MPN. We identified NFIB (JAK2V617F heterozygosity) leads to an ET phenotype. In as a gene upregulated in our MPN meta-analysis, which support of this model, we observed that JAK2V617F homo- includes a recently published data set of CD34 þ cells of PV zygosity was associated with PV (64%; 9/14) and PMF (36%; patients compared with normal CD34 þ specimens.24 In our 5/14) but not ET patients (Supplementary Table S1). SNP array study, we show that NFIB, which lies 9.1MB In addition to the role of JAK2V617F allele dosage, we centromeric to the JAK2 locus, is amplified as a result of trisomy hypothesized that specific genomic aberrations may have a 9 in 6/87 MPN patients and gain of 9p23–13.3 in 1/87 MPN role in disease phenotype. To test this, we performed patients. NFIB expression was not significantly changed follow- unsupervised hierarchical clustering of SNP profiles for ing JAK2 inhibition in HEL cells and modestly upregulated MPN patients heterozygous (allele burden range 1–85%) for following JAK2 inhibition in UKE-1 cells, which are both the JAK2V617F allele (n ¼ 63) (Supplementary Figure S3). homozygous for the JAK2V617F mutant allele. NFIB expression Although genomic instability profiles of these MPN samples was not altered by overexpression of either WT or mutant JAK2 did not segregate according to disease subtype, we found that in TF-1 cells, which express wild-type JAK2 (Supplementary PMF patients were more likely to have large (410 Mb) Figure S4). In contrast, PIM1 was modulated in response to JAK2 chromosomal aberrations compared with ET or PV patients inhibition or overexpression in keeping with its positive (P ¼ 0.009) (Figure 3). In addition, ET and PMF patients had regulation by JAK/STAT signaling. Taken together, these data

Blood Cancer Journal Identiﬁcation of novel lesions and pathways in MPN KL Rice et al 8 0.4 PV (n=23) 0.2 0.0 0.2 0.4 PMF (n=15) 0.2 0.0 0.2 Frequency 0.4 ET (n=25) 0.2 0.0 0.2 0.4 1 2 3 4 5 6 7 8 9 101112131415 171920x Chromosome

PV PMF ET P value Whole Chromosomal Aberrations/person 0.2 0.2 0.04 0.4 Regions (>10Mb)/person 0.5 2.1 0.8 0.009 Regions (1-10Mb)/person 0.6 1.8 2.0 0.01 Regions (<1Mb)/person 2.4 3.5 3.8 0.4

Figure 3 PMF and ET patients display higher frequencies of chromosomal aberrations compared with PV. Frequency plots of chromosomal gains (red) or loss (green) in individuals with PV (n ¼ 23), PMF (n ¼ 15) and ET (n ¼ 25). Y axis represents the frequency of copy number change in each patient subgroup. T-test was used to identify signiﬁcant differences in the frequency of chromosomal aberrations between these three patient subtypes.

suggest that aberrant JAK2 signaling does not account for the The most frequent aberrations observed in our patient cohort upregulation of NFIB in MPN. involved gains of whole chromosome 9 or 9pLOH, suggesting To explore the putative oncogenic potential of NFIB,we that genes on this chromosome may provide a selective clonal utilized the erythropoietin-dependent Ba/F3-EPOR murine advantage. Previous studies have shown that wild-type JAK2 is hematopoietic cell line. Stable overexpression of NFIB in capable of inhibiting STAT5 activation induced by mutant Ba/F3-EPOR cells protected cells apoptosis induced by EPO JAK2V617F.6 This inhibition is presumably due to competition withdrawal (Figure 4a) Furthermore, NFIB-infected Ba/F3-EPOR between the two JAK2 forms for cytokine receptor binding cells displayed a modest, but consistent increase in cell because wild-type JAK2 does not affect autophosphorylation number compared with vector (Po0.02, Figure 4b). Taken of JAK2V617F.8 Since JAK2 is present on chromosome 9, together, these data suggest that ectopic NFIB expression may we hypothesized that gain of chromosome 9 may provide cooperate with aberrant JAK2 signaling to dysregulate normal an alternative mechanism for cells to acquire an extra copy of hematopoiesis. mutant JAK2. In support of this, we observed that a majority of patients trisomy 9 were associated with high JAK2V617F allele burden, suggesting that JAK2 was mutated in this Discussion extra chromosome 9. Although the presence of other disease-associated genes on chromosome 9 may also The recent use of genome-wide methods to interrogate copy contribute to the pathogenesis of MPN, these data support the number alterations and gene expression changes in hematologic hypothesis that acquisition of an additional JAK2V617F mutant malignancies has been critical to our understanding of the allele, either by 9pLOH or trisomy 9, confers a selective, mechanisms underlying these heterogeneous diseases. In this competitive edge. study, we employed a combination of high-density SNP arrays The finding that JAK2 and STAT5B are significantly upregu- and gene expression arrays to identify novel oncogenic lesions lated in patients with trisomy 9 and 9pLOH, which are and dysregulated pathways in a cohort of 87 MPN patients. In comprised largely of PV patients (57%), reinforces the role of addition to identifying copy number variations affecting genes aberrant JAK/STAT signaling in this disease subtype. MAPK14 TET2, ASXL1, EZH2 and CBL, which have previously been (also known as p38a), which is also upregulated in patients with reported in MPN,34 we identified new MPN-associated aberra- chromosome 9 abnormalities, is a kinase that lies downstream of tions, which may have a role in disease initiation and/or JAK2, and is activated in response to pro-inflammatory stimuli progression. and hematopoietic growth factors. Recent studies reveal a role

Blood Cancer Journal Identiﬁcation of novel lesions and pathways in MPN KL Rice et al 9 100 90 80 EF 70 NFIB 60 50 40

Dead cell % 30 20 ** 10 0 1 U/ml 0.1 U/ml 0.01 U/ml EPO conc

9 8 EF NFIB

) 7 6 6 5 4 3 Live Cells (x10 2 1 0 12345678910 Days

Figure 4 NFIB protects cells from cytokine withdrawal-induced cell death and promotes cell growth (a). Baf3/EPOR cells transduced with control (EF) or EF-NFIB lentivirus for 48 h and GFP þ cells were replated in RPMI with 1 U/ml, 0.1 U/ml and 0.01 U/ml EPO for 48 h. Cell viability was determined by Trypan blue exclusion (b). Cumulative cell counts were performed over 10 days. Data are expressed as the average of three independent experiments±s.e.m. (**Po0.05).

for MAPK14 in red blood cell production, since mice lacking homozygous for JAK2V617F. Indeed, we found that all patients MAPK14 are anemic due to failed definitive erythropoiesis.35 harboring a gain of chromosome 17q leading to amplification of Given the association of high-level JAK2V617F expression with SOCS3, had intermediate JAK2V617F allele burden (average a PV-like phenotype in murine models of MPN,36 it is tempting 55.4%), supporting a model in which overexpression of SOCS3 to speculate that aberrant expression of MAPK14 in patients with may cooperate with JAK2V617F to enhance myeloproliferation. two copies of JAK2V617F may activate an erythroid differentia- Although aberrant JAK2 signaling is clearly a recurrent theme tion pathway, contributing to the development of PV. in MPN, it is likely that other JAK2V617F-independent events In addition to gaining extra copies of JAK2V617F, modulation contribute to disease pathogenesis. Recent studies have identi- of JAK2 kinase activity may also have a role in disease fied several disease-associated mutations such as TET2 and progression. The suppressor of cytokine signaling (SOCS) ASXL1 that may initiate a premalignant clone, since they appear proteins are negative regulators of cytokine signaling and are to precede the JAK2V617F mutation.2,11 In this study, we show expressed after cytokine stimulation, directly interacting with that NFIB, which we previously reported to be overexpressed in JAK/STAT pathway members to modulate the responsiveness of CD34 þ cells from nine PV patients compared with normal cells to further signaling.37 Although it is postulated that SOCS controls, is amplified in six patients via gains of whole proteins are tumor suppressors, there is also evidence to suggest chromosome 9 and in one patient by gain of region that overexpression of SOCS may favor tumor progression. 9p23–13.3. Previous studies reporting overexpression of NFIB Persistent expression of SOCS3 is observed in AML and in a small subset of PV patients (two with 9pLOH and two coincides with constitutive activation of JAK-STAT pathways.38 without), hypothesized that this was a consequence of Interestingly, our study identified SOCS3 within a region of 9pLOH.40 It was later shown, however, that NFIB was chromosomal gain on chromosome 17 in 6/87 patients and overexpressed in eight MPN patients positive for endogenous SOCS3 was also upregulated B2-fold in our meta-analysis. erythroid colony formation, however, none of these patients had Although SOCS3 is a potent negative regulator of erythropoietin 9pLOH.41 Although the ectopic NFIB expression may be a signaling through the EPO receptor and wild-type JAK2, a study consequence of trisomy 9, it is also possible that NFIB may be by Hookham et al.,39 revealed that SOCS3 is unable to upregulated by other factors at the level of gene transcription. negatively regulate mutant JAK2V617F and in contrast, Indeed, we observed that only two genes present in the 9pLOH enhanced the proliferation of Ba/F3 cells in the presence of region, JAK2 and C9orf46, which encodes a transmembrane EPOR. Furthermore, SOCS3 degradation was inhibited by protein with unknown function, were upregulated in these JAK2V617F and this correlated with SOCS3 tyrosine phosphor- patients compared with patients with normal chromosome 9 ylation, a feature also observed in the PBMC of patients cytogenetics. NFIB encodes a CAAT box-binding transcription

Blood Cancer Journal Identification of novel lesions and pathways in MPN KL Rice et al 10 factor and previous studies showed that overexpression in 32D transformation in patients with MPN-blast phase, suggesting the cells increases resistance to TGF-b1 inhibition.40 NFIB expres- existence of a pre-JAK2V617F clone. sion was not positively regulated by JAK2 in cell lines, however, Although unsupervised clustering of PV, ET and PMF patients overexpression in Ba/F3-EPOR cells conferred a modest growth did not reveal distinct aberration patterns associated with these advantage, and impaired apoptosis induced by EPO withdrawal. MPN subtypes, we noted that PMF patients displayed more The role of the related NFIA gene in erythropoiesis is well chromosomal aberrations that PV or ET patients. These established, and a recent study revealed structural alterations in observations are consistent with other studies showing that NFIA in 2% of chronic myeloid diseases, theoretically leading to patients with PMF exhibit increased copy number changes and inactivation of the protein.42 The functional role of NFIB in abnormal cytogenetics compared with PV or ET.17 The normal and malignant hematopoiesis remains unknown. Pre- consequences of this increased genomic instability remains liminary experiments suggest that NFIB is expressed at low unclear, however, a recent study investigating the chromosomal levels in human CD34 þ cells, however, expression is abnormalities of patients with leukemic transformation of MPN, upregulated during normal megakaryocte differentiation (data revealed that time from diagnosis to leukemic transformation not shown). Future studies will seek to understand the effect of was shorter for PMF than ET or PV patients.48 Further analysis on NFIB overexpression on normal hematopoiesis and in the a larger cohort may help us determine the aberrations associated context of the JAK2V617F overexpression. with a specific MPN subtype. In addition to NFIB, overlap of genes identified by our SNP In conclusion, we have used a combination of high-resolution array and MPN meta-analysis identified several other potentially SNP arrays and gene expression arrays to identify novel interesting disease candidates. For example, MLLT3 and potential disease candidates that may have a role in MPN. SMARCA2, both of which are located on chromosome 9 and Future studies will seek to address the specific contribution of upregulated in MPN patients, were identified as part of a gene aberrated genes and signaling pathways identified in this study set expressed by ‘normal’ quiescent HSCs.31 MLLT3 is a positive to the development of MPN. regulator of erythroid and megakaryocyte differentiation during normal hematopoiesis43 and is also implicated in leukemo- genesis, as a translocation fusion partner of MLL in AML.44 Conflict of interest EZH1, which is located in a region of chromosome gain (17q21.1–21.31) in 5/87 patients and is upregulated in the MPN The authors declare no conflict of interest. meta-analysis, is an H3K27 methyltransferase that forms part of a non-canonical PRC2 complex and has a role in Acknowledgements preserving H3K27me3, preventing derepression of PRC2 targets 45 involved in ESC pluripotency. Recent studies identified loss of We thank David Ronald, director of Genotyping Facility at Albert function mutations in EZH2 in MDS and MPN, suggesting a Einstein College of Medicine, and Greg Khitrov, director of Life tumor-suppressor activity. Although the functional conse- Science Technology Laboratory of Department of Medicine at quences of EZH1 overexpression in MPN remains to be Mount Sinai School of Medicine for carrying out Affymetrix SNP elucidated, these studies highlight an emerging theme in array experiments. We also thank Dr Weiguo Zhang (Duke U). myeloid malignancies involving the dysregulation of genes Supported by NIH Grant: R01 HL082950. involved in epigenetic regulation, such as ASXL1, UTX, SETD2, JARID1C and TET2. Recent studies have demonstrated that JAK2V617F may affect References genomic stability by displacing HP1a from chromatin46 and by deregulating DNA double-strand break repair mechanisms.47 1 Delhommeau F, Jeziorowska D, Marzac C, Casadevall N. Plo et al.,47 showed that overexpression of JAK2V617F and to a Molecular aspects of myeloproliferative neoplasms. Int J Hematol lesser extent wild-type JAK2 in Ba/F3-EPOR cells, led to an 2010; 91: 165–173. increase in spontaneous homologous recombination and 2 Tefferi A. Novel mutations and their functional and clinical relevance in myeloproliferative neoplasms: JAK2, MPL, TET2, RAD51 nuclear foci formation. JAK2V617F expression was ASXL1, CBL, IDH and IKZF1. Leukemia 2010; 24: 1128–1138. associated with genetic instability, as assessed by the higher 3 Ihle JN, Gilliland DG. Jak2: normal function and role in occurrence of centrosome abnormalities and sister chromatid hematopoietic disorders. Curr Opin Genet Dev 2007; 17: 8–14. exchange. Furthermore, increased homologous recombination 4 Liu KD, Gaffen SL, Goldsmith MA. JAK/STAT signaling by cytokine and RAD51 formation was observed in patient CD34 þ cells receptors. Curr Opin Immunol 1998; 10: 271–278. from JAK2V617F-positive PV and PMF patients.47 In support of 5 Morgan KJ, Gilliland DG. A role for JAK2 mutations in myeloproliferative diseases. Annu Rev Med 2008; 59: 213–222. these studies, we observed significantly more genomic aberra- 6 James C, Ugo V, Le Couedic JP, Staerk J, Delhommeau F, Lacout C tions in patients with intermediate JAK2V617F allele burden et al. A unique clonal JAK2 mutation leading to constitutive (1–85%) compared with patients with wild-type JAK2 (o1% signalling causes polycythaemia vera. Nature 2005; 434: JAK2V617F) (data not shown). Unexpectedly, however, patients 1144–1148. with high JAK2V617F allele burden (485%) displayed fewer 7 Kralovics R, Teo SS, Buser AS, Brutsche M, Tiedt R, Tichelli A et al. aberrations. The ability of mutant JAK2 to induce hyper- Altered gene expression in myeloproliferative disorders correlates with activation of signaling by the V617F mutation of Jak2. Blood recombination is thought to facilitate mitotic recombination 2005; 106: 3374–3376. 47 that may lead to LOH and account for disease heterogeneity. 8 Levine RL, Wadleigh M, Cools J, Ebert BL, Wernig G, Huntly BJ Whether genomic instability induced by JAK2V617F facilitates et al. Activating mutation in the tyrosine kinase JAK2 in leukemic transformation, however, remains contentious. For polycythemia vera, essential thrombocythemia, and myeloid example, in a study of 17 patients with JAK2V617F-positive metaplasia with myelofibrosis. Cancer Cell 2005; 7: 387–397. MPN that later transformed to AML, 9 patients went on to 9 Zhao R, Xing S, Li Z, Fu X, Li Q, Krantz SB et al. Identification of an 10 acquired JAK2 mutation in polycythemia vera. J Biol Chem 2005; develop JAK2V617F-negative AML. Furthermore, a recent 280: 22788–22792. 48 study by Thoenssin et al., found no significant association 10 Theocharides A, Boissinot M, Girodon F, Garand R, Teo SS, between JAK2V617F mutation status and time to leukemic Lippert E et al. Leukemic blasts in transformed JAK2-V617F-

Blood Cancer Journal Identification of novel lesions and pathways in MPN KL Rice et al 11 positive myeloproliferative disorders are frequently negative for the 30 Carbuccia N, Murati A, Trouplin V, Brecqueville M, Adelaide J, JAK2-V617F mutation. Blood 2007; 110: 375–379. Rey J et al. Mutations of ASXL1 gene in myeloproliferative 11 Delhommeau F, Dupont S, Della Valle V, James C, Trannoy S, neoplasms. Leukemia 2009; 23: 2183–2186. Masse A et al. Mutation in TET2 in myeloid cancers. N Engl J Med 31 Forsberg EC, Passegue E, Prohaska SS, Wagers AJ, Koeva M, Stuart JM 2009; 360: 2289–2301. et al. Molecular signatures of quiescent, mobilized and leukemia- 12 Gelsi-Boyer V, Trouplin V, Adelaide J, Bonansea J, Cervera N, initiating hematopoietic stem cells. PLoS One 2010; 5: e8785. Carbuccia N et al. Mutations of polycomb-associated gene ASXL1 32 James C, Ugo V, Casadevall N, Constantinescu SN, Vainchenker W. in myelodysplastic syndromes and chronic myelomonocytic A JAK2 mutation in myeloproliferative disorders: pathogenesis and leukaemia. Br J Haematol 2009; 145: 788–800. therapeutic and scientific prospects. Trends Mol Med 2005; 11: 13 Green A, Beer P. Somatic mutations of IDH1 and IDH2 in the 546–554. leukemic transformation of myeloproliferative neoplasms. N Engl J 33 Scott LM, Scott MA, Campbell PJ, Green AR. Progenitors Med 2010; 362: 369–370. homozygous for the V617F mutation occur in most patients with 14 Tefferi A, Mesa RA, Schroeder G, Hanson CA, Li CY, Dewald GW. polycythemia vera, but not essential thrombocythemia. Blood Cytogenetic findings and their clinical relevance in myelofibrosis 2006; 108: 2435–2437. with myeloid metaplasia. Br J Haematol 2001; 113: 763–771. 34 Tefferi A. Novel mutations and their functional and clinical 15 Steensma DP, Tefferi A. Cytogenetic and molecular genetic aspects relevance in myeloproliferative neoplasms: JAK2, MPL, TET2, of essential thrombocythemia. Acta Haematol 2002; 108: 55–65. ASXL1, CBL, IDH and IKZF1. Leukemia 2010; 24: 1128–1138. 16 Gangat N, Strand J, Lasho TL, Finke CM, Knudson RA, Pardanani A 35 Tamura K, Sudo T, Senftleben U, Dadak AM, Johnson R, Karin M. et al. Cytogenetic studies at diagnosis in polycythemia vera: Requirement for p38alpha in erythropoietin expression: a role for clinical and JAK2V617F allele burden correlates. Eur J Haematol stress kinases in erythropoiesis. Cell 2000; 102: 221–231. 2008; 80: 197–200. 36 Tiedt R, Hao-Shen H, Sobas MA, Looser R, Dirnhofer S, Schwaller J 17 Tefferi A, Sirhan S, Sun Y, Lasho T, Finke CM, Weisberger J et al. et al. Ratio of mutant JAK2-V617F to wild-type Jak2 determines Oligonucleotide array CGH studies in myeloproliferative neo- the MPD phenotypes in transgenic mice. Blood 2008; 111: plasms: comparison with JAK2V617F mutational status and 3931–3940. conventional chromosome analysis. Leuk Res 2009; 33: 662–664. 37 Croker BA, Kiu H, Nicholson SE. SOCS regulation of the JAK/STAT 18 Langemeijer SM, Kuiper RP, Berends M, Knops R, Aslanyan MG, signalling pathway. Semin Cell Dev Biol 2008; 19: 414–422. Massop M et al. Acquired mutations in TET2 are common in 38 Schuringa JJ, Wierenga AT, Kruijer W, Vellenga E. Constitutive myelodysplastic syndromes. Nat Genet 2009; 41: 838–842. Stat3, Tyr705, and Ser727 phosphorylation in acute myeloid 19 Mullighan CG, Goorha S, Radtke I, Miller CB, Coustan-Smith E, leukemia cells caused by the autocrine secretion of interleukin-6. Dalton JD et al. Genome-wide analysis of genetic alterations in Blood 2000; 95: 3765–3770. acute lymphoblastic leukaemia. Nature 2007; 446: 758–764. 39 Hookham MB, Elliott J, Suessmuth Y, Staerk J, Ward AC, 20 Klampfl T, Harutyunyan A, Berg T, Gisslinger B, Schalling M, Vainchenker W et al. The myeloproliferative disorder-associated Bagienski K et al. Genome integrity of myeloproliferative JAK2 V617F mutant escapes negative regulation by suppressor of neoplasms in chronic phase and during disease progression. Blood cytokine signaling 3. Blood 2007; 109: 4924–4929. 2011; 118: 167–176. 40 Kralovics R, Guan Y, Prchal JT. Acquired uniparental disomy of 21 Levine RL, Belisle C, Wadleigh M, Zahrieh D, Lee S, Chagnon P chromosome 9p is a frequent stem cell defect in polycythemia et al. X-inactivation-based clonality analysis and quantitative vera. Exp Hematol 2002; 30: 229–236. JAK2V617F assessment reveal a strong association between 41 Kralovics R, Buser AS, Teo SS, Coers J, Tichelli A, van der Maas AP clonality and JAK2V617F in PV but not ET/MMM, and identifies et al. Comparison of molecular markers in a cohort of patients a subset of JAK2V617F-negative ET and MMM patients with clonal with chronic myeloproliferative disorders. Blood 2003; 102: hematopoiesis. Blood 2006; 107: 4139–4141. 1869–1871. 22 Olshen AB, Venkatraman ES, Lucito R, Wigler M. Circular binary 42 Bernard F, Gelsi-Boyer V, Murati A, Giraudier S, Trouplin V, segmentation for the analysis of array-based DNA copy number Adelaide J et al. Alterations of NFIA in chronic malignant myeloid data. Biostatistics 2004; 5: 557–572. diseases. Leukemia 2009; 23: 583–585. 23 Beroukhim R, Lin M, Park Y, Hao K, Zhao X, Garraway LA et al. 43 Pina C, May G, Soneji S, Hong D, Enver T. MLLT3 regulates early Inferring loss-of-heterozygosity from unpaired tumors using high- human erythroid and megakaryocytic cell fate. Cell Stem Cell density oligonucleotide SNP arrays. PLoS Comput Biol 2006; 2:e41. 2008; 2: 264–273. 24 Berkofsky-Fessler W, Buzzai M, Kim MK, Fruchtman S, Najfeld V, 44 Iida S, Seto M, Yamamoto K, Komatsu H, Tojo A, Asano S et al. Min DJ et al. Transcriptional profiling of polycythemia vera MLLT3 gene on 9p22 involved in t(9;11) leukemia encodes a identifies gene expression patterns both dependent and indepen- serine/proline rich protein homologous to MLLT1 on 19p13. dent from the action of JAK2V617F. Clin Cancer Res 2010; 16: Oncogene 1993; 8: 3085–3092. 4339–4352. 45 Shen X, Liu Y, Hsu YJ, Fujiwara Y, Kim J, Mao X et al. EZH1 25 Guglielmelli P, Zini R, Bogani C, Salati S, Pancrazzi A, Bianchi E mediates methylation on histone H3 lysine 27 and complements et al. Molecular profiling of CD34+ cells in idiopathic myelo- EZH2 in maintaining stem cell identity and executing pluripo- fibrosis identifies a set of disease-associated genes and reveals the tency. Mol Cell 2008; 32: 491–502. clinical significance of Wilms’ tumor gene 1 (WT1). Stem Cells 46 Dawson MA, Bannister AJ, Gottgens B, Foster SD, Bartke T, Green 2007; 25: 165–173. AR et al. JAK2 phosphorylates histone H3Y41 and excludes 26 Gnatenko DV, Cupit LD, Huang EC, Dhundale A, Perrotta PL, HP1alpha from chromatin. Nature 2009; 461: 819–822. Bahou WF. Platelets express steroidogenic 17beta-hydroxysteroid 47 Plo I, Nakatake M, Malivert L, de Villartay JP, Giraudier S, Villeval dehydrogenases. Distinct profiles predict the essential thrombo- JL et al. JAK2 stimulates homologous recombination and genetic cythemic phenotype. Thromb Haemost 2005; 94: 412–421. instability: potential implication in the heterogeneity of myelo- 27 Tenedini E, Fagioli ME, Vianelli N, Tazzari PL, Ricci F, Tagliafico E proliferative disorders. Blood 2008; 112: 1402–1412. et al. Gene expression profiling of normal and malignant CD34- 48 Thoennissen NH, Krug UO, Lee DH, Kawamata N, Iwanski GB, derived megakaryocytic cells. Blood 2004; 104: 3126–3135. Lasho T et al. Prevalence and prognostic impact of allelic 28 D’Andrea AD, Yoshimura A, Youssoufian H, Zon LI, Koo JW, imbalances associated with leukemic transformation of Philadel- Lodish HF. The cytoplasmic region of the erythropoietin receptor phia chromosome-negative myeloproliferative neoplasms. Blood contains nonoverlapping positive and negative growth-regulatory 2010; 115: 2882–2890. domains. Mol Cell Biol 1991; 11: 1980–1987. 29 Stegelmann F, Bullinger L, Griesshammer M, Holzmann K, This work is licensed under the Creative Commons Habdank M, Kuhn S et al. High-resolution single-nucleotide polymorphism array-profiling in myeloproliferative neoplasms Attribution-NonCommercial-Share Alike 3.0 Unported identifies novel genomic aberrations. Haematologica 2010; 95: License. To view a copy of this license, visit http://creativecommons. 666–669. org/licenses/by-nc-sa/3.0/

Supplementary Information accompanies the paper on Blood Cancer Journal website (http://www.nature.com/bcj)

Blood Cancer Journal